A wide variety of synthetic resins have been developed for use as electrical insulating material, particularly material which is satisfactory for use as slot insulation in dynamoelectric machines and for use as insulation for conductors which are to be employed as magnet wires (insulated electrical conductors) in electrical apparatus. It is well known that insulating material which is to be employed for these purposes must be able to withstand extremes of mechanical, chemical, electrical and thermal stresses. Thus, wires to be employed as coil windings in electrical apparatus are generally assembled on automatic or, semi-automatic coil winding machines which, by their very nature, bend, twist, stretch and compress the enameled wire in their operation. After the coils are wound, it is common practice to coat them with a varnish solution containing solvents such as ketones, alcohols, phenols and substituted phenols, aliphatic and aromatic hydrocarbons, halogenated carbon compounds, and the like. Insulating coatings on magnet wire must be resistant to these solvents.
In order to conserve space in electrical apparatus, it is essential that the individual turns which make up the coils be maintained in close proximity to each other. Because of the closeness of the turns, and the fact that there may be a large potential gradient between adjacent turns, it is necessary that the coating resins employed as wire enamels have a high dielectric strength to prevent short circuiting between adjacent coils. In operation of electrical apparatus containing coiled wires, high temperatures are often encountered and the enamels must be able to withstand these high temperatures as well as the mechanical stresses and vibrations encountered in electrical apparatus so that the enamel coating does not soften or come off the wire.
It is also well known that the power output of motors and generators can be increased a great deal by increasing the current density in the magnet wires of these machines. However, increasing the current density through magnet wires is accompanied by an attendant rise in the operating temperature of the magnet wires. This increased temperature has meant that conventional water based organic enamels, particularly the economically attractive polyester based resins, could not be used in high current density windings because the higher operating temperatures encountered caused deterioration or decomposition of the enamel coating.
In the past, many attempts have been made to prepare magnet wires which met all of the mechanical, chemical, electrical and thermal requirements of high temperature magnet wire while still being economically feasible. Cost per unit of power output of a resulting dynamoelectric machine is a very important factor in any magnet wire insulation, since an excessive magnet wire cost tends to make a magnet wire impractical for use regardless of its properties. While excessive cost of a magnet wire is generally the result of five factors, a sixth factor, that of ecology and environmental considerations in connection with the use of organic solvents, is now of prime importance.
The first, and the most obvious, factor is the cost of the raw materials in the resin which is to be applied to the conductor. The second factor is related to the ability of the resinous material to be dissolved in readily available, inexpensive solvents. Since resinous materials are preferably stored and transported in solution, the bulk and weight of the solvent play a large part in the cost of having the resin at the place where it is to be used at the time it is to be used. In practice, it has been found that it is desirable to employ resinous materials as wire enamels which are capable of being held in solutions which contain at least 30 to 50 percent, by weight, of solids. Since the solvents in the resinous solution are generally allowed to escape without recovery from the wire coating operation, the cost of the solvent is an important factor in the cost of the cured enamel.
The third factor which vitally affects the cost of an enameled wire is the time required to cure the enamel once it has been applied to the conductor. If this time is excessive, an unduly large baking oven is required or the speed of the wire through the oven must be maintained at an uneconomically low rate. The fourth factor which affects the cost of a magnet wire is the flexibility of the conditions which may be employed in applying the resin to the conductors and in curing the resin once it has been applied. If the wire speed range in the curing operation, the curing temperature, and the wire diameter sizes are critical, it is obvious that a large amount of defective magnet wire may be prepared under mass production conditions; however, if large variations in curing conditions are tolerable, only a very small amount of the magnet wire prepared need be discarded because of defective insulation.
The fifth factor which is important in the cost of a magnet wire is the ability of the same resinous solution to be applied to both round and rectangular conductors and to conductors made of various metals. If different resin solutions must be used for each type of conductor, the time required to change the resin solution is an integral part of the magnet wire cost.
The sixth factor, which is important to the overall production process, as well as to the environment in which the production takes place, is the ecological and pollution factor, and the related safety and toxicity considerations. Organic solvents are becoming increasingly valuable, and production communities are becoming more concerned about the quality of life and the environment surrounding the manufacturing operation. Thus, it is highly important for a variety of reasons to avoid discharging and wasting organic solvents directly into the atmosphere. A related consideration with respect to the use of organic solvents is therefore the cost of handling and disposal. It has been established that for a typical organic solvent coating operation in a conventional wire tower, more than 90% of the fuel bill is utilized to heat air to dilute evaporated solvent and thereby dilute it to a nonflammable state and to burn the offgases to CO.sub.2 and H.sub.2 O before they are emitted into the atmosphere.
At the present time, commercially available coating materials for use in electrical applications, such as the coating materials disclosed in U.S. Pat. No. 2,936,296, issued May 10, 1960, to F. M. Precopio and P. W. Fox for "Polyesters From Terephthalic Acid, Ethylene Glycol and a Higher Polyfunctional Alcohol," and used and sold commercially under the trademark "ALKANEX" by General Electric Company, are widely used, highly successful and effective compositions, but have the economic and ecological disadvantage of requiring the use of organic solvents. Where organic solvents are used, they are driven off during curing of the coatings and are generally not economically recoverable. Many such solvents are becoming economically, ecologically and environmentally prohibitive, making it increasingly desirable to utilize substantially water based solvents.
A wide variety of aqueous polyester coating solutions are known in the art. With few exceptions, however, the coatings produced from such aqueous solutions are not suitable for electrical applications, particularly for wire enamels. Polyester coatings from aqueous solutions cure only very slowly to a tack-free state, exhibit excessive weight loss on curing as compared to organic solvent based resins, and, on aging, become brittle, darken, lose flexibility and generally depolymerize under the conditions of most electrical applications.
Polyamide and polyimide coating materials in aqueous solutions, and coatings produced therefrom, are generally well known in the art, and are highly effective for producing stable electrical grade coatings. See, for example, U.S. Pat. No. 3,652,500, issued Mar. 28, 1972, to M. A. Peterson, for "Process For Producing Polyamide Coating Materials By End Capping"; U.S. Pat. No. 3,663,510, issued May 16, 1972, to M. A. Peterson, for "Process For Producing Polyamide Coating Materials"; U.S. Pat. No. 3,507,765, issued Apr. 21, 1970, to F. F. Holub and M. A. Peterson, for "Method For Electrocoating A Polyamide Acid"; U.S. Pat. No. 3,179,614, issued Apr. 20, 1965, to W. H. Edwards, for "Polyamide Acids, Compositions Thereof, And Process For Their Preparation"; U.S. Pat. No. 3,179,634, issued Apr. 20, 1965, to W. M. Edwards, for "Aromatic Polyimides And The Process For Preparing Them"; and U.S. Pat. No. 3,190,856, issued June 22, 1965, to E. Lavin, et al. for "Polyamides From Benzophenonetetracarboxylic Acids And A Primary Diamine." The prior art involves generally the preparation of a coating medium containing a high molecular weight polyamide acid, and application of the coating medium to a substrate to provide a polyamide acid coating thereon, followed by the curing of the high molecular weight polyamide acid to a polyimide. While such coating materials produce coatings having desirable properties, particularly for electrical applications, they are relatively more expensive than polyester type coating materials.
Aqueous base polyamide acid systems, as described in the above-mentioned patents to Peterson, result in excellent high temperature, electrical grade coatings (250.degree. C., 40,000 hr., insulation coatings), which are stable, and easily made and used, but are relatively expensive when compared to the polyester compositions. It should be noted that the polyester (Alkanex type) magnet wire coating provides a thermal insulation barrier which, though it is less than that of polyimide magnet wire coating, nevertheless is highly suitable for a large segment of the magnet wire needs in the industry, particularly for class B applications, (135.degree. C., 20,000 hr. coatings).
Aqueous based acrylic systems, of the type described in U.S. Pat. No. 2,787,603, issued Apr. 2, 1957, to P. F. Sanders for "Aqueous Coating Compositions and Substrates Coated Therewith," while inexpensive, are not generally suitable for high temperature electrical grade coatings applications such as class B applications. Moreover, such aqueous based acrylic systems are emulsions and not solutions, thereby creating certain stability problems.
Efforts have been made to mix various emulsion polymerized resins to upgrade coatings produced therefrom. For example, properties of coatings and films from polyacrylic polyester resins in aqueous solvents have been somewhat improved by the addition of water soluble phenol-formaldehyde resins, epoxy resins and melamine resins. Such polymer blends, however, are generally not sufficiently upgraded to the classical polyester grade insulations presently utilized in the magnet wire industry.
Because of the high latent heat of vaporization of water, it is desirable in water based systems, particularly for application as wire enamels, to utilize as high a solids content as is possible, commensurate with workable viscosities, when the medium is used with automatic coating apparatus such as wire towers. High molecular weight polymers, such as the polyamide acid polymers which are described in the patents listed above, produce extremely viscous solutions except in relatively low solids content systems. For many applications, the low solids content systems are quite suitable. For wire tower use, however, the low solids content aqueous solution creates production problems which reduces the efficiency of the tower.
Criteria for electrical insulating materials, such as magnet wire insulations, slot insulations, insulating varnishes and the like have been established in the art. In order to determine whether the insulation on a magnet wire will withstand the mechanical, chemical, electrical and thermal stresses encountered in winding machines and electrical apparatus, it is customary to apply the resin to a conductor, by a method which will be described hereinafter, and to subject the enameled wire to a series of tests which have been designed to measure the various properties of the enamel on the wire.
Various tests, which will be described in detail later, include the abrasion resistance tests, the 25 percent elongation plus 3X flexibility test, the snap elongation test, the 70-30 solvent resistance test, the 50--50 solvent resistance test, the dielectric strength tests, the flexibility after heat aging test, the heat shock test, the cut-through temperature test, and the high temperature dielectric strength loss test. The enamel on a conductor which will withstand the mechanical, chemical and electrical stresses encountered in magnet wire applications and which is operable at temperatures of at least 135.degree. C. for extended periods of time must withstand at least 10 strokes with the average of three tests being not less than 20 in the repeated scrape abrasion resistance test, must withstand 980 "grams to fail" in the unidirectional scrape resistance test, must pass the 25 percent elongation plus 3X flexibility test, must show no surface defects in the snap test, must show no attack on the insulation in either of the solvent resistance tests, must have a dielectric strength of at least 1500 v. per mil twisted pair, must show no surface defects when wound on a 3X mandrel after heat aging for 100 hours at 175.degree. C., must show no defects when a 5X coil is aged for 30 minutes at 155.degree. C. in the heat shock test, and must have a cut-through temperature of at least 215.degree. C. under a 1000 gram load for 18 AWG heavy coated insulated magnet wire on copper conductor. In addition, for the same type of magnet wire with Nylon overcoat the insulated conductor must not show a loss in dielectric strength of more than 2/3 of original dielectric strength or a minimum of 1500 volts per mil twisted pair, must show no surface defects when a 3X coil is aged for 30 minutes at 155.degree. C. in the heat shock test, and must have a cut-through temperature of at least 200.degree. C. under a 1000 gram load.
The abrasion resistance tests, flexibility test, and snap test are employed to determine the mechanical properties of a magnet wire. Abrasion resistance is a measure of the amount of abrasion an insulated electrical conductor will withstand before the insulating enamel is worn away from the conductor. Repeated scrap abrasion resistance is measured by rubbing the side of a loaded round steel needle back and forth across the surface of an insulated electrical conductor until the enamel is worn away. The number of strokes required to wear the enamel away is referred to as the number of abrasion resistance strokes. Unidirectional scrape resistance is measured by rubbing the side of a round steel needle across the surface of an insulated electrical conductor under increasing load until the conductor is exposed. The load required to expose the conductor is referred to as the "grams-to-fail" load. For a complete description of the procedure followed in abrasion resistance testing where a needle is rubbed back and forth across the insulated electrical conductor, reference is made to NEMA Standard Section MW 24 which describes the procedure followed in the present invention. This NEMA Standard is incorporated by reference into the present application.
The flexibility of the enamel on a magnet wire is determined by stretching the enameled conductor and examining the stretched portion of the wire under a binocular microscope at a magnification of ten to determine if there are any imperfections on the surface of the enamel. The imperfections which may be noted on the surface of the enamel are a series of parallel surface lines of fissures which are perpendicular to the long axis of the wire. This condition of the enamel film is known as crazing. Another defect which can sometimes be observed is a break in the enamel film in which the two sections of the film are actually physically separated and the opening extends in depth to the exposed conductor. This defect is called a crack. A third defect which may be noted is a mar or blemish in the enamel film.
In the 25 percent elongation plus 3X flexibility test an insulated electrical conductor having a diameter X is elongated 25 percent and wound about a mandrel having a diameter 3X. If examination of the enamel under a magnification of ten shows none of the surface defects noted above, the insulation on the conductor passes this flexibility test. In some of the examples which follow, flexibility tests using elongations other than 25 percent and mandrels having a diameter other than 3X are employed. However, in all of these cases the flexibility test is as severe as the 25 percent elongation plus 3X flexibility test.
The snap elongation test measures the ability of the insulation to withstand a sudden stretch to the breaking point of the conductor. The insulation on the conductor must not show any cracks or tubing beyond three test wire diameters on each side of the fracture after the insulated conductor is jerked to the breaking point at the rate of 12 to 16 feet per second.
Solvent resistance tests are conducted to determine whether a magnet wire will satisfactorily withstand the chemical stresses found in electrical applications, i.e., whether the enamel is resistant to the solvents commonly employed in varnishes which may be used as an overcoat for the enameled wires. The solvent resistance test is the determination of the physical appearance of an enameled wire after immersion in a refluxing bath of a specified solution. Two solution systems are used for each sample of wire. Both of these solutions contain a mixture of alcohol and toluene. The alcoholic portion is composed of 100 parts by volume of U.S.P. ethanol and 5 parts by volume of C.P. methanol. One solvent test solution (which is designated as 50--50) consists of equal parts by volume of the above alcohol mixture and of toluene. The second solution (which is designated as 70-30) is 70 parts of the alcohol mixture and 30 parts of toluene.
In the usual operation of the test, about 250 ml. of the solution is placed in a 500 ml. round-bottomed, single-necked flask which is heated by a suitable electrical heating mantle. A reflux condenser is attached to the flask and the solution is maintained at reflux temperature. A sample is formed so that three or more straight lengths of the wire having cut ends can be inserted through the condenser into the boiling solvent. After five minutes the wire is removed and examined for blisters, swelling or softening. Any visible change in the surface constitutes a failure. Soft (requiring the thumbnail to remove it) but smooth and adherent enamel is considered to pass this five minutes test. The samples are then returned to the solvent for another five minutes and re-examined for the same defects. If the enamel shows any blisters or swelling at the end of either the five minutes or the ten minutes test in the 70-30 solution (70-30 solvent resistance test) the enamel has failed the solvent resistance test. If the enamel shows any blisters or swelling at the end of the five minutes test in the 50--50 mixture (the 50--50 solvent resistance test) the enamel has failed this solvent resistance test.
The dielectric strength of the enamel film determines whether the insulation on a magnet wire can withstand the electrical stresses encountered in electrical apparatus. The dielectric strength of an insulating film is the voltage required to pass a finite current through the film. In general, dielectric strength is measured by increasing the potential across the insulating film at a rate of 500 volts per second and taking the root mean square of the voltage at which the finite current flows through the film as the dielectric strength.
The type of specimen employed to measure dielectric strength is a sample made up of two pieces of enameled wire which have been twisted together a specified number of times while held under a specific tension. A potential is then placed across the two conductors and the voltage is increased at the rate of 500 volts per second until a finite current flows through the insulation. The voltage determined by this method is referred to as "dielectric strength, volts (or volts per mil), twisted pair." The number of twists and the tension applied to the twisted wire is determined by the size of the bare conductor. A complete listing of the specifications for various wire sizes are described in the aforementioned NEMA Standard Section MW 24.
In order to determine whether a magnet wire may be employed at high temperatures, it is necessary to measure properties of the enameled conductor at high temperatures. Among the properties which must be measured are the cut-through temperature of the enamel, the flexibility of the enamel after heat aging at an elevated temperature, the heat shock characteristics of the enamel, and the dielectric strength loss of the enamel when heated at high temperatures in air. Since it is well known that copper is the most common conductor, all of the thermal tests of magnet wire are conducted on copper magnet wire.
The cut-through temperature of the enamel film is measured to determine whether the insulation on a magnet wire will flow when the wire is raised to an elevated temperature under compressive stress. The cut-through temperature is the temperature at which the enamel film separating two magnet wires, crossed at 90 degrees and supporting a given load on the upper wire, flows sufficiently to establish electrical contact between the two conductors. Since magnet wires in electrical apparatus may be under compression, it is important that the wires be resistant to softening by high temperature so as to prevent short circuits within the apparatus. The tests are conducted by placing two eight inch lengths of enameled wire perpendicular to each other under a load of 1000 grams at the intereaction of the two wires. A potential of 110 volts A.C. is applied to the end of each wire and a circuit which contains a suitable indicator such as a line recorder, a buzzer or neon lamp is established between the ends of the wires. The temperature of the crossed wires and the load is then increased at the rate of 3 degrees per minute until the enamel softens sufficiently so that the bare conductors come into contact with each other and cause the indicator to signal a failure. The temperature at which this circuit is established is measured by a thermocouple extending into a thermowell to a point directly under the crossed wires. The cut-through temperature is taken as the temperature in the thermowell at the moment when the current first flows through the crossed wires. Although this temperature is always somewhat lower than the true temperature of the wires, it gives a fairly accurate measurement of the cut-through temperature range of the enameled wire being tested. Magnet wires designated for operating temperatures of at least 135.degree. C. should have a cut-through temperature of at least 175.degree. C.
When measuring properties of an insulating film such as percent elongation after heat aging, heat shock, weight loss after heating in vacuum, and dielectric strength loss after heating in air, what is actually being measured is the effect of thermal degradation of the enamel on the particular properties being measured. The most straightforward method of measuring this thermal degradation of an enamel on a wire is to maintain the enameled wire at the temperature at which it is desired to operate the wire until decomposition takes place.
However, this method is impractical in the evaluation of new materials because of the relatively long periods of time involved. Thus, it might be found that an enameled wire may operate successfully at a temperature of 135.degree. C., for example, for five to ten years before any substantial thermal degradation takes place. Because it is obviously impractical to wait such a long period of time to find out whether a resin is satisfactory for magnet wire enamel, it is customary to conduct accelerated heat life tests on these enameled wires. Since thermodynamic theories show that the rate of a given reaction can be determined as a function of temperature, it is possible to select elevated temperatures for thermal tests of enamel films and to calculate the thermal properties of the enameled wire at the desired operating temperature from these accelerated test data. Although it might be expected that degradation reactions which occur at elevated test temperatures might not occur at temperatures at which the magnet wire is to be operated because of activation energies required to initiate certain reactions, experience has shown that accelerated heat life tests are an accurate method for determining the heat life of a material at operating temperatures.
In determining whether an enamel film will lose its flexibility after extended periods of time at operating temperature, it is customary to heat age a sample of the enameled wire. In practice it has been found that for a magnet wire to be satisfactory for use in dynamoelectric machines at temperatures of at least 135.degree. C. a sample of the enameled wire having a conductor diameter X must show no surface defects when wound on a mandrel having a diameter of 3X after heat aging for 100 hours in a circulating air even maintained at a temperature of 175.degree. C.
The effect of high temperatures on the flexibility of a magnet wire enamel may also be measured by winding a sample of the enameled wire having a conductor diameter X on a mandrel having a diameter of 5X, removing the sample of wire from the mandrel and placing it in a circulating air oven maintained at 155.degree. C. After 30 minutes the sample of wire should show no surface defects in any of the windings in order for the enameled wire to have sufficient flexibility for steady operation at least 135.degree. C. This test is known as the heat shock test.
The final thermal requirement of a magnet wire which is to be used at elevated temperatures is that the dielectric strength of the enamel film remains sufficiently high at elevated temperatures after a long period of operation so that no short circuits occur between adjacent magnet wires. We have found that for a magnet wire to be satisfactory for operation at a temperature of at least 135.degree. C. its dielectric strength should not be less than two-thirds of the initial dielectric strength after being maintained at a temperature of 175.degree. C. for 100 hours in an oven circulating air having a relative humidity of 25 percent at room temperature. This change in dielectric strength is measured as the dielectric strength, volts (or volts per mil) twisted pairs, both before and after the 175.degree. C. heat aging.